© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 27689
    Prev
    Next
    INDEX
    Explanations

    instances of the word "widely."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    74
    +0.12
    0.7%
    421
    +0.10
    0.5%
    326
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    74
    +0.12
    0.01
    421
    +0.10
    0.01
    450
    +0.09
    0.01
    Negative Logits
    ts
    -1.61
    atar
    -1.58
    !”
    -1.55
    mers
    -1.54
     whole
    -1.52
     sooner
    -1.51
     entire
    -1.50
    ando
    -1.45
    iken
    -1.42
    rill
    -1.41
    POSITIVE LOGITS
    usta
    1.64
    zzo
    1.48
     disseminated
    1.47
    ust
    1.44
    ]{.
    1.41
    dorff
    1.39
     dispersed
    1.39
     equ
    1.39
     circulated
    1.38
     distribut
    1.34
    Activations Density 0.062%

    No Known Activations